Lead Data Scientist - AML

Description

The Anti-Money Laundering (AML) Risk team builds systems that combine unsupervised and supervised learning with GenAI to detect and reduce financial crime worldwide.

The work supports the growing number of customers, transactions and products at the company.

Responsibilities

• Build AML detection controls that mix unsupervised, semi-supervised and supervised learning with GenAI.
• Build frameworks that show control coverage at regional level.
• Develop technology that works for the company's diverse international users.
• Work with colleagues across functions to own and solve AML problems at global scale.
• Mentor junior team members on both technical and non-technical skills.
• Evaluate AML systems against internal and external benchmarks.
• Build decisioning layers that find the best balance between precision and recall.
• Give data-driven insight into likely outcomes under different scenarios.
• Work with operations teams to improve processes and feed their feedback into automation.
• Run projects that use spare operational capacity, such as manual data labelling to improve models.
• Build systems that give investigators detailed insight into red flags and typologies on profiles and transactions.
• Package algorithms into deployable libraries or objects and move them from staging to production.
• Set up and maintain scheduled data collection and model retraining through automated pipelines.
• Maintain production-grade Python services.

Requirements

• Experience implementing, training, testing and evaluating the performance of machine learning systems.
• Strong Python skills. Experience with object-oriented programming principles is a major advantage.
• Experience with statistical analysis and designing well-structured experiments.
• A strong product mindset and the ability to work independently across functions and teams.
• Good communication skills, with the ability to explain points clearly to non-technical people.
• Strong problem-solving skills, including helping refine problem statements and working out how to solve them.
• Useful but not essential: experience automating operational processes with technical solutions, such as large language models.
• Useful but not essential: willingness to sit in on operational side-by-sides to understand their pain points.
• Useful but not essential: knowledge of or experience in the financial crime domain.

Conditions

Equal opportunity: the company welcomes applicants from all backgrounds and encourages applications from under-represented groups.